DUTIES REGARDING ANIMALS

Patrick Kain

[Forthcoming in Kant’s Metaphysics of Morals: A Critical Guide.

ed. Lara Denis. Cambridge University Press]

Abstract: A better appreciation of Kant’s commitments in a variety of disciplines reveals Kant

had a deeper understanding of human and non-human animals than generally recognized, and
this sheds new light on Kant’s claims about the nature and scope of moral status and helps to
address, at least from Kant’s perspective, many of the familiar objections to his notorious
account of “duties regarding animals.” Kant’s core principles about the nature of moral
obligation structure his thoughts about the moral status of human beings and non-human animals.
Kant’s commitments in biology, psychology, anthropology and physical geography support his
account of the nature of and distinction between humans and non-human animals. This account
supports Kant’s judgment that we have duties to every human being and significant duties
regarding non-human animals, duties which involve direct concern for animals because of their
nature. A comparison of Kant’s account with some recently proposed Kantian alternatives
provides additional perspective on some of the distinctive features, and strengths and
weaknesses, of Kant’s approach.
 1

CHAPTER ELEVEN: DUTIES REGARDING ANIMALS

PATRICK KAIN

INTRODUCTION

In one of the most widely cited, and certainly the most criticized, passages from the

Metaphysics of Morals, Kant infamously insists, in part:

As far as reason alone can judge, a human being has duties only to human beings

(himself and others), since his duty to any subject is moral constraint by that

subject’s will. Hence the constraining (binding) subject must, first, be a person;

and this person must, second, be given as an object of experience, since the

human being is to strive for the end of this person’s will and this can happen only

in a relation to each other of two beings that exist … But from all our experience

we are acquainted with no being other than a human being that would be capable

of obligation (active or passive). A human being can therefore have no duty to

any beings other than human beings; and if he represents to himself that he has

such duties, it is because of an amphiboly in his concepts of reflection, and his

supposed duty to other beings is only a duty to himself. He is led to this

misunderstanding by mistaking his duty with regard to other beings for a duty to

those beings. (MS 6:442)1

Kant insists that we have duties to all human beings: “a human being is under obligation to

regard himself, as well as every other human being, as his end” (MS 6:410). But “a human being

1
Translations in this essay are those of the Cambridge Edition, except where noted otherwise, or
where a quoted passage is not included in an already-published Cambridge Edition work. Here
I have slightly modified Gregor’s translation of “kennen”and “sich vorstellen” to better capture
technical epistemological features of the amphiboly.
 2

has duties only to human beings (himself and others)” (MS 6:442). While Kant recognizes many

moral constraints upon our behavior toward non-human animals, he insists that these are only

duties “with regard to these animals,” rather than duties “to those beings” (MS 6:442-443).2

“Every human being has a legitimate claim to respect from his fellow human beings and is in

turn bound to respect every other”(MS 6:462). But animals are “things,” not persons, and

“respect is always directed only to persons, never to things” (KpV 5:76). This makes it seem as

if an animal is no more worthy of our concern than is a turnspit on which we might choose to

roast it.

This position on the nature and scope of fundamental “moral status” and its practical

implications both for the treatment of so-called “marginal cases” of seriously immature or

radically disabled human beings and for the treatment of non-human animals has been a source

of much consternation.3 Prominent philosophers have suggested that one of Kant’s greatest

mistakes überhaupt was his failure to appreciate the nature of non-human animals and their

moral significance.4 Kant is regularly accused of (i) drawing an arbitrary distinction between the

moral status of all human beings and that of non-humans which cannot be reconciled with the

actual condition of human infants and severely disabled adults, (ii) a fundamental failure to

consider the nature of non-human animals and acknowledge their similarity to humans, (iii) a

failure to recognize that the moral constraints on human behavior toward non-humans should be

based on the nature of those animals, rather than in incidental effects of our behavior upon

2
In what follows, I will often use “animals” as shorthand for “non-human animals.”
3
“To have moral status is to be morally considerable, or to have moral standing. It is to be an
entity towards which moral agents have, or can have, moral obligations.” Mary Anne Warren,
Moral Status: Obligation to Persons and Other Living Things (Oxford University Press, 1997),
3.
4
Responses by Christine Korsgaard and Peter Singer in Vadim Vasilyev’s “International Kant
Interview 2004-2005,”
www.philos.msu.ru/community/staff/vasiliev/Kant_Interview/Kant_Interview.htm
 3

humans which turn upon highly contingent features of human psychology, and (iv) a failure to

regard animals as the proper objects of human concern in their own right.5 These charges appear

to cut to the heart of Kant’s ethics and addressing them has seemed to demand either the outright

rejection of Kantian ethics or significant alteration of its trademark focus on human dignity. 6

A better appreciation of Kant’s commitments in a variety of disciplines reveals Kant had

a deeper understanding of human and non-human animals than generally recognized and this

sheds new light on Kant’s claims about the nature and scope of moral status, helping to address,

at least from Kant’s perspective, many of the familiar objections to his notorious account of our

“duties regarding animals.” In section one, I will review some of Kant’s core principles about the

nature of moral obligation which structure his thoughts about the moral status of human beings

and non-human animals. In section two, I will consider in some detail Kant’s account of the

nature of and distinction between humans and non-human animals. With this account in hand, I

will turn, in section three, to Kant’s case for claiming that we have duties to every human being

5
For example, Alexander Broadie and Elizabeth M. Pybus, “Kant’s Treatment of Animals,”
Philosophy 49 (1974), 375-83; Peter Singer, Animal Liberation: A New Ethics for Our
Treatment of Animals (New York: Random House, 1975); Tom Regan, The Case for Animal
Rights (London: Routledge & Kegan Paul, 1984); Martha Nussbaum, Frontiers of Justice:
Disabilities, Nationality, Species Membership (Cambridge, Mass.: Harvard University Press,
2006); and Warren, Moral Status. Elements of these criticisms can be found in the work of
Arthur Schopenhauer and Albert Schweitzer, see Heike Baranzke, “Tierethik, Tiernatur und
Moralanthropologie im Kontext von §17 Tugendlehre,” Kant-Studien 96 (2005), 336-63.
6
For significant concessions by Kantians on some of these points, see Christine M. Korsgaard,
The Sources of Normativity, ed. Onora O'Neill (Cambridge University Press, 1996); “Fellow
Creatures: Kantian Ethics and Our Duties to Animals,” in Grethe B. Peterson (ed.), The Tanner
Lectures on Human Values, vol. XXV (Salt Lake City: Utah University Press, 2005), 77-110;
and “Interacting with Animals: A Kantian Account,” in Thomas Beauchamp and R.G. Frey
(eds.), The Oxford Handbook on Ethics and Animals, (Oxford University Press, 2010 [in
press]); Jens Timmermann, “When the Tail Wags the Dog: Animal Welfare and Indirect Duty
in Kantian Ethics,” Kantian Review 10 (2005), 128-49; Allen W. Wood, “Kant on Duties
Regarding Nonrational Nature I,” Proceedings of the Aristotelian Society, Supplement 72
(1998), 189-210, Kant’s Ethical Thought (Cambridge University Press, 1999), and Kantian
Ethics (Cambridge University Press, 2008).
 4

and significant duties regarding non-human animals that are grounded in their nature. Finally, in

section four, I will consider Kant’s account in relation to some recently proposed Kantian

alternatives.

1. THE BASIS OF MORAL STATUS

Kant insists upon a sharp distinction between beings with dignity (Würde) and those with

mere price (Preis). Price is a kind of relative value, a value something has if it is related in the

correct way to something else, in particular to the needs or desires of human beings. By contrast,

dignity is a kind of absolute and intrinsic value; something with dignity “is raised above all price

and therefore admits of no equivalent,” it cannot “be replaced by something else” (G 4:434).7

Kant claims that what gives a being dignity and marks it out as an “end in itself” is its

innate rational capacity (Fähigkeit) for autonomy, a predisposition (Anlage) to “personality,” the

capacity to “legislate” the moral law and to act out of respect for the moral law, “freedom…

under moral laws” (G 4:428, 435-36; MS 6:223, 418; RGV 6:27).8

A human being regarded as a person, that is, as the subject of a morally practical

reason, is exalted above any price; for as a person (homo noumenon) he is not to

be valued merely as a means to the ends of others or even to his own ends, but as

an end in itself, that is, he possesses a dignity (an absolute inner worth) by which

he exacts respect for himself from all other rational beings in the world. He can

7
This is not to deny that Kant draws some distinctions within “price.” See, for example, G 4:
428, 434; MS 6:434.
8
Kant rejects the possibility that organisms, in general, could be “final ends” or ends in
themselves (KU 5:425-35), contra G.F. Meier, Philosophische Sittenlehre (Halle: Hemmerde,
1753-1761), §975.
 5

measure himself with every other being of this kind and value himself on a

footing of equality with them. (MS 6:434-35)

In Kant’s theory there is a deep connection between dignity and moral obligation. In

Kant’s terms, only beings with dignity are capable of “passive” and “active obligation”: only

beings with dignity can be obligated or obligate others. “Duty to any subject is moral constraint

by that subject’s will” (MS 6:442). Moral obligations can be articulated as the demand to respect

the dignity and autonomy of every rational being (G 4:428–36). Thus, the second formula of the

categorical imperative demands: “So act that you use humanity, whether in your own person or

in the person of any other, always at the same time as an end, never merely as a means” (G

4:429). Lest it appear that Kant is simply begging all of the relevant questions about the scope of

moral status, we must note that Kant employs the terms “humanity and “personality” in a

technical sense to refer to certain capacities or predispositions of the will, which may or may not

turn out to be ascribable to all and only human beings.

Since a “duty to any subject is moral constraint by that subject’s will,” an obligator (a

being to whom one can have a duty, a being capable of “active obligation”) must have a will that

can impose a moral constraint upon the obligated, and the obligated (one capable of “passive

obligation”) must have a will that can be constrained by the obligator. Thus, Kant isolated two

necessary conditions for genuine moral status: we can be obligated only to a being that is both (i)

a “person,” a being with a free will “standing under the moral law” and (ii) is “given as an object

of experience,” so that we can recognize that it can obligate us and so that we can, through our

actions, have some bearing upon it and/or its ends (MS 6:442).

Regarding the first condition, Kant famously argued that neither “theoretical” philosophy

nor empirical investigation can establish that there is any such absolute freedom, any “freedom
 6

under moral laws.” “Experience lets us cognize only the law of appearances and hence the

mechanism of nature, the direct opposite of freedom” (KpV 5:29). Kant came to insist that the

reality of absolute freedom, or freedom under moral laws, can only be established in practical

philosophy, by the “fact of reason.” We are each “immediately conscious (as soon as we draw up

maxims of the will for ourselves)” of the moral law; the moral law is given to us as “the sole fact

of pure reason” and this fact leads us to the concept of freedom and the postulation of its reality

in us (KpV 5:29–31). “The categorical imperative proves for morally practical purposes” that at

least some of us “human beings” are free” (MS 6:280n).

Yet we must not neglect the second condition and its implications for the determination

of moral status. Kant’s insistence that we can only have obligations to persons who are “given as

an object of experience” suggests that experience and the biological, psychological, and

anthropological theories, concepts, and judgments, through which we make systematic sense of

the objects we are given in experience, must play a significant role in helping to determine in a

naturalistically respectable way which objects of experience should be considered to be the

presentation of the relevant kinds of predispositions; a suggestion confirmed by Kant’s appeal to

“experience” and his employment of biological and psychological terminology in the discussion

of our duties regarding non-human animals and of the moral relationship between human parents

and the children they conceive (MS 6:280, 442). We must investigate salient aspects of Kant’s

investigations in these disciplines.

 7

2. THE NATURE OF ANIMALS

Kant articulated a naturalistic framework for systematic biological and

psychological investigations.9 Kant insists that, in natural science, we must seek to

identify a system of efficient or “mechanical” causal laws responsible for observable

regularities, but there are phenomena that resist such an understanding (KU 5:387–88,

372–76, 401–04). To bring such regularities “under laws” a set of teleological concepts

are needed, including the concept of an organism, a “natural end” which is a

teleologically organized and self-organizing whole, organized for life and reproduction

(VR 2:429; KU 5:376). When using such concepts, we must still observe the maxim that

“in a natural science everything must be explained naturally” (GtP 8:178; cf. KrV

A544/B572, A773/B801). One should seek a systematic and parsimonious account which

relies upon analogies to observed powers and eschews both unnecessary and unhelpful

complexity and direct appeals to divine intervention. A “philosophically appropriate,”

“naturalistic” explanation of the regularities observed among organisms favors an

“epigenetic” theory of the reproduction of organisms combined with a commitment to

real biological species and a doctrine of original “predispositions” (GtP 8:168–69; BBM

8:102; GtP 8:178). In reproduction, adult organisms of a species produce a new organism

of their species, endowed at conception with the species’s specific organization, a set of

“predispositions” (Anlagen) and “germs” or “seeds” (Keime) that were originally

implanted in the species’ first members (KU 5:423).

9
The argument of the next several paragraphs is developed in more detail in Patrick
Kain, “Kant’s Defense of Human Moral Status,” Journal of the History of Philosophy
47 (2009), 59-102.
 8

In psychology we find an account of animals or “living” organisms, those

endowed with “sensation and choice” (MS 6:442), which extends this biological

framework. Since animals can perceive and respond to changes in their immediate

environment in ways that (most) plants cannot, Kant insists upon judging animal

behavior as a product of inner principles (even if less than fully conscious or self-

conscious ones): living beings have the capacity to move themselves according to the

power of choice, that is, in virtue of their representations. Kant argued that although

mental representations can, in general, be cognized and explained naturalistically, neither

can be fully explained “materialistically.” The mental representations that are essential

constituents of the genuine psychological regularities we observe, Kant argued, must be

regarded as states of an immaterial soul (though not necessarily a simple, substantial, or

immortal soul) (KrV B419–20; KU 5:460). Kant insisted that animals are not “mere

machines,” but have souls with a vis locomotiva, because the mental representations that

guide their behavior cannot be realized in matter (KU 5:457, 464n). In animals, the

“faculty of desire” is linked with a “faculty of cognition” or “intuition” which gives rise

to representations (via the senses, but also via reproductive and anticipatory imagination)

and a “faculty of feeling pleasure or displeasure” in conjunction with a representation

(MS 6:211).10 For systematic reasons, Kant favored an account of animal reproduction

and original ensoulment according to which each animal is endowed from its conception

with the biological and psychological predispositions of its species.11 The predispositions

and propensities of an animal species, which may underlie or manifest themselves in a

10
See also Lectures on Metaphysics [hereafter: VM] 28:115-17, 274-77, 448-49, 594,
690; 29: 906, 1026.
11
Kain, “Defense,” 82-87.
 9

variety of instincts, acquired inclinations, and habits, serve as causal grounds for the

occurrence of certain thoughts, feelings, desires, and behaviors.12

We humans can be “immediately aware” of our own representations, especially

those representations upon which we act; based on observable similarities between our

actions and the behavior of non-human animals, we infer that they have some capacities,

analogous to, if yet specifically different from, our capacity to reason and our capacity to

act from reason. To take a prominent example:

In comparing the artistic actions of animals with those of human beings,

we conceive of the ground of the former, which we do not know, through

the ground of similar effects in humans (reason), which we do know, and

thus as an analogue of reason, and by that we also mean to indicate that

the ground of the artistic capacity in animals, designated as instinct, is in

fact specifically different from reason, but yet has a similar relation to the

effect (comparing, say, construction by beavers with that by humans). (KU

5:464n)

Animals can represent, perceive and be acquainted with objects through their

representations and are capable of subtle differentiations amongst objects.13 Some

animals have more refined external senses than we humans (VM 28:277). In some cases,

12
Patrick Frierson, “Kant’s Empirical Account of Human Action,” Philosophers’ Imprint
5, no. 7 (2005), 1-34.
13
LJ 9:64-65; FS 2:59-60; PS 2:285; HN 15:161-62, 713; VM 28:66-67, 78-79, 98-99,
857. For a careful analysis of Kant’s account of the nature and limits of animal
psychology upon which I rely in this paragraph, see Steve Naragon, “Reason and
Animals: Descartes, Kant, and Mead on the Place of Humans in Nature,” unpublished
doctoral dissertation, University of Notre Dame, 1987, and “Kant on Descartes and the
Brutes,” Kant-Studien 81 (1990), 1-23. See also, Karl Ameriks, Kant's Theory of
Mind (Oxford:Oxford University Press, 1982/2000), 242.
 10

it seems “the acts of animals arise out of the same principium from which human actions

spring, and the animal actions are analogues of this” (C 27:459). We have no access

through our own introspection, however, to evidence that animals have inner sense,

concepts, or cognition which we encounter in our own case. Kant thought animals do not

possess a capacity for language use, which would indicate concepts and higher cognition,

much less a first-person pronoun. As for the observed artistry of beavers, mentioned

above, Kant endorsed Bonnet’s contention that beavers always build dams according to a

single model or plan, an indication that whatever their artistry and the complex form of

social cooperation they employ, they lack the ability to reflect upon, modify and improve

their craft or inhibit their instincts.14 More generally, animals’ behavior appears to be

guided by rather determinate and pervasive instincts; they are incapable of impulse

control and many are easily duped; their behavior does not progress cumulatively over

the course of generations. Parsimony counsels not ascribing more sophisticated mental

capacities than necessary to explain the phenomena, so Kant concluded that nothing in

their behavior required positing full-blown “consciousness,” a capacity for “inner sense,”

for second-order representations, including representations of oneself or one’s entire

condition: animals lack concepts, judgment, apperception and self-consciousness, and

thus genuine cognition of objects. Unable “to represent to themselves the ground of their

movement [Beweggründe],” they cannot reflect upon their desires or have “a desire

within a desire” (VM 28:99). Unable to conceive of “what is useful or injurious” or

14
VPG-Hesse, 122-123; see also VM 28:117. My thanks to Werner Stark for allowing
me access to his transcriptions of the notes from Kant’s “Lectures on Physical
Geography,” some of which will be included in volume 26 of the Akademie Ausgabe
(forthcoming), and his invaluable assistance in working with them. I cite passages from
these lectures by name (e.g., VPG-Hesse) and the pagination in the original
manuscripts. Translations are my own.
 11

“desirable in regard to [their] condition as a whole,” they are unable to pursue or

experience happiness as such. Perhaps most important for present purposes, absent the

capacity to represent what is “unconditionally good,” animals must lack the capacity to

act upon (or against) the representation of such an unconditional law (KrV A802/B830).

In contrast, we human beings have language, “inner sense” and second order

representations, concepts, apperception, self-consciousness, cognition, and capacities for

reflection and inhibition in light of general representations. In his Anthropology text,

Kant claimed that each of humans’ three practical predispositions, the “technical,

pragmatic and moral” predispositions, distinguish human beings from all other terrestrial

animals. The profound indeterminacy of our instincts and skills, and the connection

between our “consciousness” and our technical skill at manipulating things (especially

with our hands) itself distinguishes us from all other animals with which we are familiar;

our capacity to use other humans in pursuit of happiness and culture and to govern

ourselves according to rational principles distinguish human beings yet further (ApH

7:321ff.)15 This creates an opening, in the human case, for Kant to contend that we have

also a capacity for a rational will: to maintain that “the categorical imperative proves for

morally practical purposes” that at least some of us “human beings” are free “persons”

with the predisposition for freedom under moral laws (MS 6:280n).

It is seldom recognized that, in addition to his interest in distinguishing human

beings, and human behavior and mental capacities, from those of non-human animals in

general, Kant had a significant interest in animal ethology, comparative morphology, and

15
There is an ambiguous relationship between this description of the practical
predispositions and the description found in the Religion (animality, humanity, and
personality) (RGV 6:26-28).
 12

natural history, as part of a proper “pragmatic” knowledge of the world. Freshly

transcribed and edited notes from his lecture course on “Physical Geography” show Kant

synthesizing the observations of leading biologists and travelers into characterizations of

non-human animals that go beyond the occasional comments in his published works

(including the Physical Geography text he allowed to be published in 1802).16

On the basis of Kant’s comments in the Anthropology and the morphological

similarities between humans and monkeys (particularly the hand, so emphasized by

Linnaeus and Buffon) we might expect Kant to have had particular interest in monkeys.17

While impressed by their manual dexterity and its deployment for catching mussels,

making beds, putting on clothes, and other things, Kant was less than fully impressed,

given reports that they steal produce from field and garden and band together to slay

lions, tigers, or even humans.18

Although the monkeys have an analogon rationis, no analogon moralitatis

will be found in them, as they are always wicked, spiteful and obstinate,

and everywhere they go, they wreak havoc.19

Wickedness is [the monkey’s] primary attribute; it is never capable of

complete trust; with respect to its mental powers, so to speak, the dog and

elephant are much to be preferred.20

16
The course originally included some anthropological topics (as did the metaphysics
course); by the mid-1770s, Kant conceived of “anthropology” and “physical
geography” as complementary “pragmatic” disciplines which he then taught alternating
semesters (VR 2:443; Br. 10:146) For a brief overview in English, see Steve Naragon’s
“Kant in the Classroom” internet resource, www.manchester.edu/kant/.
17
ApH 7:322; VPG-Pillau 252, 266.
18
Kant, Physical Geography, [hereafter: PG] 9:336-37.
19
VPG-Kaehler 405; VPG-Messina 248.
 13

Indeed,

[Dogs] seem to be the most perfect animal, and to manifest most strongly

the analogon rationis… they carefully look after their responsibilities,

remain with their master; if they’ve done something wicked they become

disturbed; and if they see their master angry, try to win him over with a

submissive posture.21

While dogs may be Kant’s prime example of brutes’ necessitation per stimulos and the

lack of impulse control -- “a dog must eat if he is hungry and has something in front of

him” (C 27:267) -- Kant notes how dogs learn to howl or open a gate-latch, and how with

practice they can learn a rabbit’s tricks and outwit a rabbit. Their instinct, by repetition of

similar cases, “forms an experience which serves the dog as a guiding thread,” despite its

lack of concepts.22

Kant’s greatest sense of wonder, though, is reserved for elephants. “When one

observes their strength and their similarity to man, [an elephant] is an animal worthy of

admiration [ein bewunderungswürdiges Thier].”23 The elephant’s trunk is “the most

noble tool,” comparable to a hand in its dexterity and sensitivity, and with a wider range

of uses as well; an elephant can use its powers more generally than any other animal.24

Elephants are very useful, because of their strength and speed on land and in water, and

20
VPG-Pillau 266. The comparison of beavers, monkeys, dogs and elephants seems to
have been a common trope, see for example Buffon’s discussion in his volume on
Elephants.
21
VPG-Kaehler 401-402. On faithfulness to their master, see also VPG-Hesse 117; MS
6:443; C 27:459.
22
VM 29:949; 28:116; Kant, Lectures on Anthropology, VA 25:1196.
23
VPG-Pillau 252. The most detailed discussion is found in the parallel Pillau and Barth
notes. These are the primary source for the rest of this paragraph, unless otherwise
noted.
24
VPG-Kaehler 397; cf. PG 9:328.
 14

because they are teachable (gelehrig) and prudent (klug). “Unprovoked, an elephant does

no one harm.”25 “It is often so gentle that one can break coconuts open on its head,

although it must be given some or it will avenge itself with its trunk.” They may not only

be tamed, but also “disciplined” (perhaps the only animal that is capable of discipline).

Kant notes that people in Surinam use an elephant in place of a servant, a role which they

carry out well and patiently.26 In one set of notes, Kant is reported to have concluded his

comments on elephants thus:

An elephant is a gentle animal, and seems to be an Analogon of Morality.

It understands jokes, but cannot be duped.27

Unfortunately, neither the precise basis of such remarks, nor their implications, are

further elaborated. Clearly the reports of elephant behavior (or at least the parts that Kant

found credible or worth collecting and remarking upon) made a significant impression

upon Kant. Rather than emphasize differences between or the distance between elephants

and humans, Kant attributes significant mental sophistication to elephants and uses words

with significant positive ethical overtones (prudence, good-natured, patience, discipline)

without reservation.28

This survey of Kant’s systematic, “naturalistic,” and empirical biology, empirical

psychology, and pragmatic anthropology and “physical geography” establishes that Kant

25
cf. PG 9:329.
26
VPG-Kaehler 396; cf. VPG-Messina 238
27
VPG-Pillau 253. This is an important contrast with most other animals, which Kant
thinks are easily deceived (VM 28:116).
28
Aside from distinguishing discipline from mere learning, Kant does not elaborate. He
appears to accept the myth that elephants do not mate in captivity, but does not mention
Buffon’s interpretation of this as a form of modesty or self-control (VPG-Pillau 235).
Nor does he elaborate an interpretation of elephant’s desire to avenge itself (when a
coconut is not shared) or resist being duped.
 15

had a serious account of the nature of animals. While many of the details and

assumptions of Kant’s account have been superceded by subsequent scientific and

philosophical developments, it is not clear that his primary conclusions have been.29 Kant

concluded, as a contingent empirical matter, that human beings have rational souls, while

no other animals of which we are familiar do.30 This is the account to which his moral

philosophy makes reference.

3. MORAL IMPLICATIONS

3.1. Kant’s Defense of Human Moral Status

As we have seen, on Kant’s account, moral status requires the possession of

“freedom under moral laws” by something “given as an object of experience.” Kant

claims that, in our “immediate consciousness” of the moral law “the categorical

imperative proves for morally practical purposes” that at least some of us “human

beings” are “free” (KpV 5:29; MS 6:280n). Of course humans infants and the severely

disabled fail to manifest in their behavior much complex consciousness at all, much less

an immediate consciousness of the moral law.31 In response to charges of arbitrariness in

Kant’s ascription of moral status to such so-called “marginal cases,” I have argued

29
Interestingly, Korsgaard seems independently to arrive at some similar conclusions in
“Interacting.”
30
See note 10, above, and Kain, “Defense,” 82n70. For additional historical context, see
Hans Werner Ingensiep, “Tierseele und tierethische Argumentationen in der Deutschen
philosophischen Literatur des 18. Jahrhunderts,” NTM: Internationale Zeitschrift für
Geschichte und Ethik der Naturwissenschaften, Technik und Medizin (N.S.) 4 (1996),
103-118.
31
It is sometimes supposed that Kant’s claims about “personality” suggest a “Lockean”
approach to personal identity and moral status, but this is dubious. Kain, “Defense,”
65n16.
 16

elsewhere that Kant has a principled basis for his ascriptions.32 First, Kant’s analysis of

freedom contends that freedom must be an original and essential predisposition of any

being that can possess it. Kant insisted that it must be possible for finite beings endowed

with freedom to come into being, since “the categorical imperative proves for morally

practical purposes” that at least some of us “human beings” are free, but he argues that it

is logically impossible for free beings to be the product of any physical operation. Kant

suggested that the most appropriate way to think about the origins of a human being is

that rational souls are created endowed with freedom and that these souls are embodied or

“brought over into the world” by human parents when they generate and ensoul a human

organism (MS 6:280).

Second, Kant’s thesis about freedom as a predisposition, taken in conjunction

with Kant’s biological, psychological and anthropological commitments, provides

support for his judgment that every human being possesses it. Kant’s biological theory

maintains that each organism can be considered the presentation of a being with

predispositions, and we must consider them to be such presentations “from procreation”

or conception. Kant’s psychological theory maintains that each animal gets its soul at the

point of its reproductive origin. The practical doctrine of original freedom entails that free

rational souls must be essentially free rational souls, which implies that moral status

attaches as soon as an organism endowed with such a soul is generated or conceived. The

patterns of pragmatic and moral development across human populations strongly

suggested to Kant that the predisposition to personality should be considered a

predisposition of the human species, as opposed to a predisposition of only some of its

32
Kain, “Defense,” 90-100.
 17

members. “The human procreative faculty is the faculty of a human being, with a human

of the other sex, to put a person in the world” (HN 23:357). Kant’s commitments provide

a principled, if debatable, basis for his judgment that all human beings, even the

apparently “marginal cases,” are intrinsically worthy of respect and each is capable of

directly obligating us. Kant’s substantive judgments about human marginal cases may

not require the rejection or radical revision of his account of moral status.

3.2. Kant’s rejection of duties “to” animals

In this context, the question becomes whether careful attention to the nature and

behavior of any non-human animals provides evidence that it, and by extension the other

members of its species, possess the predisposition to personality. Kant’s conclusion was

that they do not. Indeed, his judgment was that there was insufficient evidence to even

ascribe to non-human animals many of predispositions and capacities which are

necessary components of the predisposition to personality: they lack the capacity for

concepts, self-consciousness, judgment, and so forth. While it is not clear precisely why

he interpreted the behavior of monkeys, dogs, and elephants as he did, absent the

manifestation by some of those animals of rather full-blown “Kantian” moral

consciousness, or at least the manifestation that such consciousness was developing, this

judgment is hardly arbitrary. Animals are “endowed with sensation and choice” yet are

“non-rational,” they are incapable of rational cognition and, most importantly, they lack a

free rational will (MS 6:442-443). Love, fear, admiration, and amazement are proper for

a variety of objects, especially for animals, but the “proper object of respect” is the moral

law and those beings with dignity, ourselves and other human beings, with the capacity to
 18

“legislate” the law and to hold it before us (KpV 5:76-78; MS 6:443; G 4:435-36, 440).33

This is why we cannot have any duties to animals.

Perhaps what strikes many readers as fundamentally objectionable about Kant’s

denial of duties to animals is the apparent implication that they are completely devoid of

moral significance, mere “things” at best only accidentally distinguishable from any

arbitrary hunk of matter. But before jumping to such a conclusion, careful attention must

be paid to the details of Kant’s positive account of the place of animals in the moral life.

3.3. Kant’s account of duties “regarding” animals

In the Metaphysics of Morals and in notes from his “Lectures on Ethics,” Kant

identifies a general duty to oneself to refrain from unjustified “violent and cruel treatment

of animals,” as well as a number of more particular moral requirements regarding our

behavior towards certain animals. After laying out Kant’s core argument for this general

duty and considering the basis for some of the particular duties he mentions, we will

examine some important and illuminating objections to it.

Kant’s contention in the Metaphysics of Morals is that the fundamental moral

problem with “violent and cruel treatment of animals” is its rather “intimate opposition”

to “a human being’s duty to himself” (MS 6:443). As Baranzke has recently emphasized,

Kant’s discussion of duties regarding animals comes at the conclusion of his discussion

of perfect duties to oneself, before he proceeds to his detailed examination of imperfect

duties to oneself or any duties to others.34 In the Metaphysics of Morals, duties to oneself

are tied to the ethical requirement to have “one’s own perfection” as an end (MS 6:385-

33
See also, KU 5: 372, 482n; C 27:459; V 27:709-710.
34
Baranzke, “Tierethik.”
 19

387). The perfect or limiting or “negative duties [to oneself] forbid a human being to act

contrary to the end of his nature and so have to do merely with his moral self-

preservation.” In contrast, positive, widening, imperfect duties to oneself “command him

to make a certain object of choice his end, concern his perfecting of himself… they

belong to his cultivation (active perfecting) of himself” (MS 6:419). Suicide, for

example, is contrary to one’s perfect duty to oneself because of the way it conflicts with

the agent’s natural inclination to self-preservation; it involves “renouncing his

personality” and “debasing humanity in [his] person” (MS 6:420, 422-23). Because of

what the agent expresses about his nature when he violates a perfect duty to himself, such

actions are particularly dishonorable.

Kant contends, most fundamentally, that the “violent and cruel treatment of

animals” violates a perfect duty to oneself. As Denis has explained, Kant insists that “the

ways that we treat animals reflect and affect morally important attitudes and feelings.”35

This approach emphasizes two points: one about the moral significance of certain of our

feelings, the other about the nature of animals and how, given that nature and our own,

animals must engage these feelings. First, “certain emotional predispositions are

extremely useful natural tools for us as moral beings,” useful both motivationally and

epistemically, and they “may also reflect certain moral commitments” insofar as they

“can be shaped” by our choices.36 In particular, Kant singles out the “disposition of

sensibility… to love something … even apart from any intention to use it” and especially

the “natural predisposition” to the “shared feeling of [other’s] suffering” as feelings that

35
Lara Denis, “Kant’s Conception of Duties Regarding Animals: Reconstruction and
Reconsideration,” History of Philosophy Quarterly 17 (2000), 405-23, 417.
36
Denis, “Kant’s Conception,” 406-407.
 20

may “promote morality or at least prepare the way for it” and are “very serviceable to

morality in one’s relations with other people” (MS 6:443).

We might go even further, once we note that Kant recognizes some “feelings,”

namely “moral feeling, conscience, love of one’s neighbor [die Liebe des Nächsten,

Menschenliebe], and respect for oneself (self-esteem),” as “moral endowments” that “lie

at the basis of morality, as subjective conditions of receptiveness to the concept of duty”

(MS 6:399). Although these are not feelings one could have a duty to have (both because

their presence is necessary for beings like us and a precondition of duty itself, and

because they may not be produced or increased, either ex nihilo or simply at will);

nonetheless, these feelings ought to be cultivated, and more importantly in the present

context, they ought not be degraded, demeaned, or devalued. The two feelings to which

Kant directly appeals in his discussion of duties regarding animals and inanimate nature,

love and sympathy, are intimately connected with feelings on this list. The general

capacity for love as “delight,” (Liebe des Wohlgefallens, amor complacentiae) “pleasure

joined immediately to the representation of an object’s existence” is discussed as part of

Kant’s treatment of the “moral endowment” of Menschenliebe, which is itself either a

special instance or a particular development of this type of feeling (MS 6:402, 449, 450).

Although sympathy does not itself appear explicitly on the list of aesthetic preconditions

of duty, it seems to have a similar status. “Sympathetic joy and sadness (sympathia

moralis) are sensible feelings of pleasure or displeasure… at another’s state of joy or

pain” (MS 6:456). Humans, Kant claims, have a natural receptivity to such shared

feeling, often called humanity or humaneness (Menschlichkeit, humanitas aesthetica),

which is a precondition for the willingness to share in other’s feelings. “While it is not in
 21

itself a duty to share the sufferings (as well as the joys) of others, it is a duty to

sympathize actively in their fate; and to this end it is therefore an indirect duty to

cultivate the compassionate natural (aesthetic) feelings in us” (MS 6:457). At least for

“animals endowed with reason,” such as ourselves, sympathetic feeling is a necessary

precondition for moral obligation. In other words, these feelings are not simply morally

useful, as merely one means among others or merely useful because of some highly

contingent facts about human psychology; they may be “an essential part of the

fulfillment of duty itself,” at least for beings anything much like us.37 These are feelings

which we have a perfect duty to ourselves to preserve and neither denigrate nor demean,

in addition to be feelings which we have an imperfect duty to ourselves to cultivate.

The second crucial point in Kant’s case for this perfect duty to ourselves

regarding animal cruelty is that, on Kant’s account of the nature of animals, animals by

their nature properly engage our morally significant feelings. An animal is not only a

beautiful and teleologically organized creature, but also a creature that can feel pleasure

and pain, that can represent the world and have desires (including desires conducive to its

self-preservation, reproduction, and enjoyment), and that can act upon those desires and

“principles” analogous to ours. Such a creature is a proper object of our love and

sympathy in ways that plants, machines, and crystal formations are not. It is “because of

these analogies” between human and animal nature that “the ways that we treat animals

reflect and affect morally important attitudes and feelings.”38 “Many of our morally

important sentiments do not discriminate between animals and humans,” and this is no

37
Paul Guyer, Kant and the Experience of Freedom (Cambridge University Press, 1993),
390. (Guyer does not endorse this specific analysis, or the point to which I am putting
it.)
38
Denis, “Kant’s Conception,” 417.
 22

accident or psychological quirk.39 It is love and/or sympathy which we feel, or at least

have a predisposition to feel, towards animals as well as human beings, and in many

cases, such feelings may be based upon the presence of some of the same, or closely

analogous, features present in animals and humans. Choices to deny, avoid, trivialize or

cavalierly violate such bonds of love or sympathy (or predispositions to them) express

disrespect for ourselves. In general, the violent or cruel treatment of animals (at least

when unjustified), is incompatible with respect for ourselves because it essentially

involves the disregard, denial or demeaning of these predispositions, feelings, and bonds

which are integral to our own nature as moral animals. Animals ought not to be harmed

or destroyed “without reason” (C 27:459).

In the case of certain kinds of animals and particular individual animals, Kant

suggests a few additional conclusions. An animal’s specific capacities, not just for

experiencing pain but for excessive strain, or for loyalty, may come into play, as may its

individual history. The kinds of work to which an animal or kind of animal may

permissibly put should accord with their capacities, they “should not be strained beyond

their capacities” (MS 6:443). Horses and dogs may provide service over many years, and

dogs in particular may do so with particular loyalty and attachment to their master, as we

have seen. Having done so, they must be rewarded with gratitude, “just as if they were

members of the household”; “once the dog can serve no longer, [we] must look after him

to the end” rather than “turn him out,” starve him, or have him shot. Failure to do so

reveals “a very small mind,” and is contrary to one’s humane or sympathetic feelings

(MS 6:443; C 27:459; V 27:710). A dog’s capacities for particular kinds of feelings,

39
Ibid., 407.
 23

desires, and attachments make it the proper object of greater love and sympathy than is

appropriate to feel for a grub, and one’s own dog’s particular devotion makes its

especially apt for a significant measure of one’s love, sympathy, and gratitude.40 One can

see how Kant’s analysis would entail similar, indeed stricter, requirements for the

treatment of elephants, given his understanding of their nature, especially their “analogy

of morality.”

Of course, it is not that feelings of love or sympathy for animals, all by

themselves, provide a rule for action. No feeling, not even “moral feeling” itself, plays

such a role in Kant’s theory, and feelings of sympathetic love, even when directed at

other humans, are neither an infallible guide to other’s needs nor by themselves a rule for

action (MS 6:400; G 4:398). Moreover, Kant explicitly allows the killing of some

animals “quickly (without pain)” and even some “agonizing physical experiments” for

important ends, though not for sport or pure speculation (MS 6:443; C 27:460). As with

other perfect duties, what needs to be determined in each domain is which courses of

action, or, better, which maxims of action are incompatible with respect for one’s rational

nature, in this case, incompatible with one’s moral self-preservation. Just as the

assumption of some risks to life and bodily integrity are compatible with the prohibition

on suicide (and with proper regard for the inclination to self-preservation); so may some

use, some killing, even some cruel treatment of animals for important human ends, be

permissible or even required.41 In The Metaphysics of Morals Kant intends to outline

40
While this is a duty to oneself, the duty requires gratitude to the dog, contra
Timmermann, “Tail,” 132; or if gratitude proper entails respect, some analogue of
gratitude to it (MS 6:454). Kant does condone killing dogs if they become rabid,
however (VPG-Hesse 117).
41
Denis, “Kant’s Conception,” esp. 413-14. See MS 6: 422-24, 434-37.
 24

some first principles that provide a basic framework for such deliberations and

determinations, rather than to provide an algorithm or exhaustive treatment of examples.

In this case, Kant’s principles may raise significant questions about a wide range of

human conduct, from animal research, to our eating and farming practices, to some of our

leisure activities; not just any human interest may justify the killing of or cruelty to an

animal.42 There are both general protections for all sentient creatures and various

particular requirements regarding specific kinds of animals and specific kinds of human-

animal relationships, requirements which depend significantly upon the nature of the

animals in question.

Before considering some objections, it is important to note how this core

argument differs from the argument often attributed to Kant. It is often thought that

Kant’s only objection to animal cruelty focuses on the putative psychological effects of

violence and cruelty toward animals on the human agent that perpetrates it, and

especially, the effects on other humans that the agent may subsequently encounter and be

more likely to mistreat. While Kant cannot resist endorsing such plausible empirical

theses, appeals to the long-term consequences of animal cruelty should not be confused

with the particularly “intimate opposition” to one’s duties to self that Kant intends to

highlight. The more familiar “brutalization argument” is vulnerable to the familiar

objection that a single act of gratuitous cruelty may fail to have discernable long-term

impact, and to the objection that the contingencies of human psychology upon which

42
For a sketch of some such arguments, see Denis, “Kant’s Conception.” Without
endorsing all of her conclusions, one can see how this approach might address a
remarkably wide-range of ethical questions. For further development of this framework
and its application to the topic of abortion see Lara Denis, “Animality and Agency: A
Kantian Approach to Abortion,” Philosophy and Phenomenological Research 76
(2008), 117-37.
 25

generalizations about long-term brutalization depend are not deep enough to properly

secure significant prohibitions on animal cruelty. But Kant’s core argument evades both

of these complaints: his focus is upon the immediate disregard for one’s morally

significant feelings that is integral to the mistreatment of animals, even in isolated

instances, and this is independent of many psychological contingencies. Still, Kant’s core

argument, as interpreted above, must confront some significant objections.

One objection focuses on Kant’s characterizations of our duties regarding animals

as “indirect” duties (MS 6:443). Timmermann has recently argued that, within Kant’s

theory, the identification of something as an indirect duty reveals that it is “really no

duty, nor part of a duty, but a mere accidental means to fulfilling a duty.” Thus, in the

case of indirect duties regarding animals, “there is nothing about the animal that makes

treating them decently morally good. Treating animals decently is a mere means to taking

care of your own moral well-being.” “There would be no duty to do it if neglect did not

lead to adverse effects on our moral capacities.”43 Surely, it is alleged, this fails to do

justice to ordinary moral intuitions about the mistreatment of animals. This objection

falters on several counts. First, in the present case it misunderstands Kant’s position:

Kant’s emphasis is upon what mistreatment of animals expresses about one’s feelings and

moral perfection, rather than on the effects of mistreatment, for oneself or another, or on

the ineffectiveness of mistreatment as a means to one’s obligatory ends. On Kant’s

account, the relevant feelings are also much more than accidental or incidental means to

moral compliance. Second, there is indeed something about the animals in question that

43
Timmermann, “Tail,” 140, 143n10, 144n11. See also Jens Timmermann, “Kant on
Conscience, ‘Indirect’ Duty, and Moral Error,” International Philosophical Quarterly
46 (2006), 293-308.
 26

grounds Kant’s demands to treat them decently: because of their nature or behavior,

animals are the proper object of one’s sympathy and love. Again, proper treatment of

animals is a necessary condition for and perhaps a constitutive part of one’s moral well-

being, rather than a mere “instrumental” means to it. Thus, regardless of how other cases

may fit Timmermann’s general characterization of “indirect” duties, Kant’s account of

our duties regarding animals does not manifest its objectionable aspects.

A second objection also focuses on the apparent “indirectness” of Kant’s account.

By focusing, as Kant’s account does, on the human agent and her own self-regarding

psychological states, it marginalizes, distorts, or attenuates the proper consideration of the

animals’ nature or proper concern for the animals and their well-being. By focusing on

the agent’s self-respect, the Kantian account seems to foreground the agent’s self-concern

(if only for her own integrity or “self-righteousness”) and background her concern for the

animals.44 Such an orientation, it is objected, is both psychologically peculiar and

ethically deficient. However, this objection may involve a confusion, at least as it is

applied to the account outlined above. Indeed, part of what Kant insists upon is the fact

that a self-respecting person is directly concerned with the fate of animals: he regards

animals as proper direct objects of love and sympathy and he acts in ways that preserve

his own disposition to such love and sympathy. To be sure, Kant will insist that one’s

love and sympathy for animals (similarly as with such feelings when had for other

humans) should, in action, be regulated by reason. But, it is not clear that self-respect

plays a larger psychological role in the case of duties regarding animals than it does in the

44
Wood, “Duties,” 194. Part of Wood’s endorsement of this objection may depend upon
his acceptance of Guyer’s claims that that the duties, on Kant’s account, must be only
imperfect, rather than perfect, duties to oneself (210n18).
 27

case of duties to other humans; rather, it is simply that there is no need to appeal, in the

present case, to the agent’s respect for anyone other than the agent. Put another way, if

the Kantian account of an agent’s self-respect leaves sufficient psychological room for

genuine respect, love and sympathy for other people when we discharge our duties to

them (and manifest love or sympathy for them), then there may be no special problem

about having direct love or sympathy for animals when we discharge our duties regarding

them.45

Understood in its proper context, Kant’s insistence upon duties to all human

beings and duties regarding animals is reasonably well-grounded and responsive to many

familiar objections. Of course, some, including some Kantians, may still insist that

animals are due greater regard than Kant allows.

4. TWO KANTIAN ALTERNATIVES

Two distinguished Kantian ethicists, Allen Wood and Christine Korsgaard, have

recently proposed modifications of Kant’s account, designed to accord animals greater

significance in Kantian ethics. It may be instructive to consider these alternatives and a

few of the ways they compare with Kant’s position as described above.

Wood rejects Kant’s claim that all duties must be duties to some person or duties

to respect rational nature “in the person of some being who has it” (and thus to respect

“persons themselves”); Wood contends, “we should also respect rational nature in the

abstract, which entails respecting fragments of it or necessary conditions of it, even

45
Consideration of the general “one thought too many” objection is beyond the scope of
this paper.
 28

where these are not found in fully rational beings or persons.”46 Some of the features of

animals (e.g., their capacity for suffering, or for desire, or for caring) constitute

“substructures, fragments, and analogues of rational nature”; they are of the sort to be

large and rather immediate components of rational nature, at least when possessed by

beings with a rational nature. Because of this special relationship these features bear to

rational nature, each instance of such a feature deserves respect in its own right.47 On this

account, Kantian duties regarding animals are established without any need for special

reference to the agent’s own self-respect (or for his respect for other human beings) and

respect is not limited to persons, even while all value is still determined in relation to

rational nature.

One point of concern about Wood’s account is that it remains unclear precisely

what “respect for rational nature in the abstract” is supposed to denote.48 More

importantly, it is unclear why the relationship that substructures, fragments, and

analogues of rational nature allegedly bear to “rational nature in the abstract” entails that

they are worthy of the genuine respect rational nature is. Indeed, precisely because they

are at best only analogues of, or tokens of a type to be components of an instance of

rational nature, one might suppose that what they are worthy of is an analogue of respect,

46
Wood, Kant’s Ethical Thought, 143; “Duties,” 193, 198.
47
Wood, Kantian Ethics, 100-103; “Duties,” 197. Occasionally Wood seems to suggest
that what deserves respect in these cases are the animals themselves, rather than the
features. Perhaps this is a further inference -- they deserve respect because they are
bearers of features that deserve respect.
48
If it is simply respect for the moral law, considered as an abstract object or principle,
that is fine, but, until it is independently determined what the moral law demands
regarding animals and their features, it cannot carry much weight in Wood’s argument.
If it is some kind of respect for the human species and its historical vocation, it may
again be unobjectionable, but would, in any event, constitute a detour much like that it
intended to avoid by eschewing appeal to the agent’s own self-respect.
 29

or some instance of a type of attitude that may be a component of respect, rather than

full-blown respect. If this is correct, one might note that the resulting view is not far from

Kant’s own view, as long as the love and sympathy of which he claims they are proper

objects are sufficiently plausible “analogues of respect.” After all, such love and

sympathy are direct forms of concern for the animals in virtue of their analogous

characteristics; and they do generate significant constraints on our behavior toward the

animals.49

Korsgaard argues that Kantians should recognize all animals (and perhaps all

functionally organized objects) as “the source of legitimate normative claims… that must

be recognized by all rational agents”; animals and their interests “have a direct normative

claim” on us, and it is their protection that the moral law demands in a fundamental or

ultimate way.50 On Korsgaard’s “constructivist” interpretation of Kant, all norms are

constructed by and all value is conferred by our acts of legislative volition.51 In pursuing

my interests, I claim that my interests and my “natural good” are worthy of pursuit by

any rational being and I claim that I possess absolute worth, worthy of respect by any

rational being; my “legislative volition” confers value upon myself and upon my interests

and constructs universal norms for my protection and the promotion of my interests. Of

course beings such as animals or human infants that are incapable of or simply fail to

exercise legislative volition ipso facto do not construct any norms or confer value on

anything. Yet, Korsgaard explains, this need not preclude someone else from constructing

49
This may be all that Wood’s position is intended to capture, since he resists ascribing
to animals any moral status equivalent to that of human beings, even most marginal
human beings or what he calls “persons in the extended sense.” Kantian Ethics, pp. 97,
101.
50
Korsgaard, “Fellow Creatures,” p. 95. (See also Korsgaard, “Interacting” and Sources.)
51
Ibid, 95, 101.
 30

norms for their protection or conferring value, even fundamental value, upon them.

Indeed, if it is my “animal nature, not just [my] autonomous nature, that [I] take to be an

end-in-itself” or what is of fundamental value, and if it is on my “natural good” as an

animal that I “confer normative value” when I value myself as an end-in-itself, then my

acts of “legislative volition” (which must be universal in scope) commit me to endorsing

the fundamental normative significance of all other humans and animals and of their

interests.52 So the moral law demands respect for animals and protection for them and

their interests, and does so for their sake.

For present purposes, we should focus on Korsgaard’s case for identifying animal

selves and an animal’s natural good as the objects of fundamental normative significance

(and proper direct objects of respect).53 First, it may be important to distinguish between

the normative significance of a particular being or self, on the one hand, and the

normative significance of that being’s interests, on the other. Even in the

straightforwardly human case, it seems important for Kantians to distinguish between

respect for a person and the concern for her interests or even her happiness or well-being

as a whole that is rooted in that respect.54 Second, it is important to recall that Kant’s own

account already requires serious concern for the interests of animals, tied to our love and

sympathy; what Kant does not allow is concern in the form of respect for the animal

52
Ibid, 104. In “Interacting” Korsgaard distinguishes “weaker” and a “stronger” versions
of this argument. As far as I can tell, to reach the conclusion that morality requires
respect for animals themselves, the stronger version is necessary.
53
For a discussion of Kant’s alleged moral constructivism, see Patrick Kain, “Self-
Legislation in Kant’s Moral Philosophy,” Archiv für Geschichte der Philosophie 86
(2004), 257-306.
54
As Korsgaard herself might say, the former has intrinsic or unconditional value while
the latter has conditional yet objective value. Creating the Kingdom of Ends
(Cambridge University Press, 1996), 260-262.
 31

itself.55 In support of the claim that it is animal nature per se upon which we confer

absolute value, Korsgaard adopts a thought experiment: “imagine that [you are] about to

be deprived of [your] rational nature, but may now settle the question whether [you] will

afterward be tortured or not. Can [you] really say: ‘In that case it won’t matter’?”56

However, even if there is agreement that it would “matter,” this thought experiment does

not isolate the precise reason for this concern, whether the reason is the same as in the

ordinary case, and this is what is needed to distinguish the accounts of Korsgaard and

Kant here.57 Does torturing “me-sans-my-rational-nature” matter because it is

disrespectful to the victim, or because it is painful, or because it is destructive? If it is

simply my love or sympathy that is, or should be, engaged in such a case, then I may not

conclude with Korsgaard that it must be “my animal self” upon which I confer absolute

value, or that, by extension, that all animals must be respected. 58

Consideration of these two alternatives puts us in a position to consider two final

points, one critical of Kant, the other complimentary. First, the criticism. Each of these

alternatives recognizes the need for a Kantian theory of value that goes, in some respects,

beyond what has been found in Kant. Wood argues for the value of substructures,

fragments and analogues of rational nature found in animals. Korsgaard appeals to a

Aristotelian account of the final ends (or natural good) of animals and insists upon the

55
This is a further reason why Korsgaard’s “stronger” argument may be required.
56
For a similar argument, see Timmermann, “Tail,” 135.
57
It is questionable whether it is metaphysically possible to be deprived of one’s nature,
or a part of one’s nature, while continuing to exist; it remains unclear whether there is a
coherent approximation appropriate for the task.
58
Other significant questions about Korsgaard’s account concern the precise nature of
this respect for animals- if it is substantially the same as that for humans- and whether
respect, perhaps equal respect, is also required for plants and machines.
 32

centrality of our animal nature to our practical identity.59 I have suggested that Kant

points in a slightly different direction, in the case at hand, namely to claims about animals

as the proper objects of human beings’ love and sympathy. But here too, Kant hardly

provided an exhaustive account of the distinctive ways in which animals properly engage

these feelings, and what he did suggest needs much more philosophical attention that it

has received. There are questions about the justification of this account within Kant’s

system and its philosophical adequacy for the tasks at hand. What a comparison of these

alternative highlights is that Kant left many unresolved questions about the nature of non-

moral value, its various species, and their precise relations to dignity, and that this

presents a challenge for Kantian accounts of duties regarding animals.

On the positive side, Kant’s warning about an “amphiboly in moral concepts of

reflection” may contain more insight than is generally appreciated. In the Critique of

Pure Reason, Kant identified an amphiboly, alleging that Leibnizian metaphysical

principles mistakenly result from a failure to distinguish properly between two different

sources of representations (namely sensibility and the understanding) (KrV A260-

292/B316-349). In moral philosophy, Kant suggests a similar confusion amongst sources

of cognition is involved when we mistake a “duty with regard to” animals “for a duty to

those beings” (MS 6:442). Carelessness with the rational concept of obligation (which

allows the thought of obligation only to persons), combined with a failure to distinguish

properly amongst our feelings, generates confusion. Our feelings of love and sympathy

do help us to “recognize… something improper” in the mistreatment of animals, but

when these feelings are not carefully distinguished from that of respect, we mistakenly

59
If I am correct about Kant’s biology and psychology, the former point may be less
foreign to Kant than Korsgaard may realize.
 33

“represent to ourselves” that we have duties to animals, even though this is contrary to

what “reason alone can judge”: such a relation cannot even be “thought” with such

beings, since they lack the relevant predisposition or capacity (V 27:710; MS 6:442).

Nonetheless, the strictness of duty and the immediacy of loving and sympathetic concern

for an animal make it feel as if we have a direct duty to the animal. Regardless of the

particular judgment about animals, given Kant’s moral philosophy and moral psychology,

some amphiboly should be expected in moral philosophy. Moral obligations will have

some implications for the treatment of beings lacking moral status, and it is only natural

that we might misinterpret our feelings in such cases as indicative of duties to such

beings. This is why Kantians, such as Kant, Wood, and Korsgaard, might be seen as

arguing amongst themselves, at least in part, about where the amphiboly occurs. If Kant

is right, those not privy to his theory of obligation and his moral psychology may be

especially vulnerable to the amphiboly, since it is hard to identify without these

philosophical resources. As it turns out, many complaints about Kant’s account of duties

regarding animals, especially those coming from theorists who reject his distinction

between respect and love or sympathy, may miss the mark because they fail to grasp the

amphiboly. While non-Kantians complain on the basis of the amphiboly that Kant

trivializes or distorts our duties regarding animals, it is not clear that Kant’s theory

demands, at a fundamental level, much less regard for non-human animals than many of

its rivals do. Indeed, his account of duties regarding animals endorses direct appreciation

of and concern for animals and recognizes significant moral requirements on us that are

grounded in and can vary with the nature, behavior, and history of the animals. If this

were all that is involved in “moral status” or a duty to something, then there might be
 34

little difference between Kant and many of his rivals. What Kant does argue for is

something more, namely respect, for human beings, and this may be something that many

of his rivals cannot accommodate. Whatever the outcome of this dispute, careful attention

to the claimed amphiboly enables us to distinguish these issues.

CONCLUSION

Examined carefully in the light of Kant’s corpus, Kant’s account of our duties

regarding non-human animals is less vulnerable to many familiar criticisms than

ordinarily thought. Perhaps Kantian theorists, if they are willing to follow Kant’s lead

and able to integrate contemporary scientific accounts of human and non-human animals

into Kantian theory, can defend the foundations of Kantian moral philosophy while both

affirming the importance of genuine concern for animals and distinguishing such concern

from respect due to human beings.